# Cluster by Keyword

When you have a lot of related nodes with no grouping, the AI can automatically analyze them and create visual boundaries that group related items by theme or keyword. It's a fast way to bring structure to a messy board.

## How to Use It

1. **Select three or more** nodes on the canvas.
2. Click the **AI Adjust** button in the selection toolbar.
3. Choose **Cluster by keyword** from the menu.

<figure><img src="/files/TayzDC5t9gAoSjjmjCTI" alt=""><figcaption></figcaption></figure>

## What Happens

The AI will:

1. Analyze the text content of the selected nodes.
2. Identify common themes or keywords.
3. Create visual group boundaries around related nodes, each with a descriptive label.

The resulting boundaries appear on the canvas just like any other group — you can rename them, resize them, or delete them afterwards if the AI's grouping isn't quite right.

## When Cluster by Keyword Works Best

* You have a long list of imported items (applications, capabilities, processes) and no idea where to start.
* The node labels are meaningful — clustering is based on text, so *"CRM System"* clusters well, *"App 1"* does not.
* You want a first-pass grouping that you'll refine manually.

## When to Use Something Else

* **You already know how you want them grouped** — just select the nodes and use [Grouping](/models/work-in-the-app/editor/grouping.md) to create a boundary manually.
* **You want connections, not boundaries** — ask [AI Chat](/models/work-in-the-app/ai/chat.md) to *"Connect related nodes with dashed lines."*
* **You only have a couple of nodes** — clustering needs at least 3 and is more useful with 10+.

## Related

* [AI Text Adjust (Toolbar)](/models/work-in-the-app/ai/text-adjust.md) – The other AI Adjust menu action
* [Grouping](/models/work-in-the-app/editor/grouping.md) – Create boundaries manually


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.ins-pi.com/models/work-in-the-app/ai/cluster.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.